Mobile Machine Sensor Calibration via Signal Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor calibration systems for mobile machines are limited in their ability to accurately calibrate sensors that detect object characteristics and require the machine to interrupt its tasks for self-testing, reducing productivity and efficiency, and are primarily designed for single-machine use.
Innovation Solution
A sensor calibration system that includes multiple machine-mounted sensors and a controller to compare and correct signals from first and second sensors, allowing for in-situ calibration without the need for the machine to be precisely positioned or for physical service, using positioning devices and calibration objects to determine the actual sensing location and apply necessary transformations to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the machine interrupts its task to perform self-testing calibration, then sensor accuracy is improved, but productivity is reduced
Solution Approach 1:
The system performs calibration actions in advance by storing reference data and transformation parameters before actual calibration is needed. The controller applies pre-computed transformation parameters to sensor data, eliminating the need for real-time mechanical repositioning or interruption of work tasks.
Solution Approach 2:
The patent replaces mechanical repositioning and physical calibration procedures with computational transformations. Instead of mechanically repositioning sensors or using physical calibration objects, the system uses software-based transformation parameters to achieve calibration, allowing calibration to occur during normal operation without mechanical interruption.
2Measurement precision
If the system requires precise positioning for calibration, then measurement accuracy is improved, but device complexity increases
Solution Approach 1:
The system creates a computational model (copy) of the sensor system's geometric relationships and uses transformation parameters to represent the calibration state. Instead of physically positioning sensors with precision, the system copies the geometric relationships into software transformations, simplifying the physical requirements while maintaining accuracy.
Solution Approach 2:
The patent changes the calibration approach from physical positioning parameters to transformation parameters. Instead of requiring precise mechanical positioning, the system uses mathematical transformation parameters (rotation, translation, scaling) that can be computed and applied software-based, reducing physical complexity while maintaining measurement accuracy.
3Device complexity
If the calibration system is designed for single-machine use, then device complexity is reduced, but adaptability is limited
Solution Approach 1:
The calibration system is designed with universal applicability across multiple machines. The controller stores transformation parameters and calibration data that can be applied to different sensors on different machines. The system can calibrate various sensor types (laser scanners, radar, cameras) on multiple mobile machines using the same fundamental approach, eliminating the need for machine-specific calibration systems.
Data Source
AI summary
A sensor calibration system for a mobile machine is disclosed. The sensor calibration system may have a first machine mounted sensor disposed on the mobile machine and configured to sense a characteristic of an offboard object and to generate a corresponding signal, and a second machine mounted sensor disposed on the mobile machine and configured to sense the characteristic of the offboard object and to generate a corresponding signal. The sensor calibration system may also have a controller in communication with the first and second machine mounted sensors. The controller may be configured to compare the characteristic of the offboard object as sensed by the first machine mounted sensor to the characteristic of the offboard object as sensed by the second machine mounted sensor, and to correct subsequent signals received from the first machine mounted sensor based on the comparison.